Certification Of Safety-critical Systems Research Articles

Bayesian Safety Validation for Failure Probability Estimation of Black-Box Systems

Estimating the probability of failure is an important step in the certification of safety-critical systems. Efficient estimation methods are often needed due to the challenges posed by high-dimensional input spaces, risky test scenarios, and computationally expensive simulators. This work frames the problem of black-box safety validation as a Bayesian optimization problem and introduces a method that iteratively fits a probabilistic surrogate model to efficiently predict failures. The algorithm is designed to search for failures, compute the most-likely failure, and estimate the failure probability over an operating domain using importance sampling. We introduce three acquisition functions that aim to reduce uncertainty by covering the design space, optimize the analytically derived failure boundaries, and sample the predicted failure regions. Results show this Bayesian safety validation approach provides a more accurate estimate of failure probability with orders of magnitude fewer samples and performs well across various safety validation metrics. We demonstrate this approach on three test problems, a stochastic decision-making system, and a neural-network-based runway detection system. This work is open-sourced (https://github.com/sisl/BayesianSafetyValidation.jl) and is currently being used to supplement the FAA certification process of the machine learning components for an autonomous cargo aircraft.

Certification of Safety-Critical Systems

Seeking new approaches toward ensuring the safety of software-intensive systems.

Self Diagnostics and Isolation Mechanisms for Mixed Criticality Systems

Virtualization is a technology that is frequently employed in computers and servers to provide isolation for execution environments, and to support the execution of multiple Operating Systems (OS) on the same hardware platform. In the embedded systems´ world, virtualization has been a rising trend, essentially because it offers an isolation mechanism that provides hardware manufacturer´ independence and it avoids obsolescence issues. The isolation mechanism supports safety and security measures, and assists in the certification of safety-critical systems. Virtualization offers improved performances, better transparency, portability and interoperability by integrating hardware and software resources, and also networking services into one computing entity. It makes the integration process of Mixed Criticality Systems (MCS) easier. For industries, FieldProgrammable Gate Arrays (FPGAs) hardware solutions provide the needed level of flexibility and performance. In this paper, a Self-test application is integrated in the hardware and also in the software level. The importance of self-test applications for Instrumentation and Control (I&C) systems will be discussed in the context of virtualization. For this implementation a type 1 hypervisor called Xtratum is used. An analysis of inter-partition communication channels´ performance will be provided including the implications multicore approaches will have on communication. The novelty of this work is to study the isolation impact multicore approaches can have on inter-partitions communications in Xtratum. Another novel aspect is the implementation of a self-test application in the hypervisor and the board as well.

Argument Representation for Dependable Computer-Based Systems

Society is becoming increasingly reliant upon the dependability of computerbased systems. Achieving and demonstrating the dependability of systems requires the construction and review of valid and coherent arguments. This paper discusses the need for a variety of classes of arguments in dependable systems and reviews existing approaches to the representation of arguments in each of these classes. The issues surrounding the certification of safety critical systems demonstrate the current need for richer representations of dependability arguments which support tools for their construction and review. The paper discusses how a meta-logical framework, informed by aspects of both formal and informal logic, offers a rich and unified means of representing dependability arguments and of thus addressing this need.

The potential for a generic approach to certification of safety critical systems in the transportation sector

Abstract This paper investigates the potential for common treatment of certification of safety critical programmable electronic systems in the transportation industries. It contains a comparative review of new, emerging international standards that are likely to influence certification procedures in the railway, automotive and aerospace sectors in the future. These include the EUROCAE/SAE aerospace guidelines, the CENELEC railway standards and IEC-61508, the draft international standard on safety related systems. The review identifies the common and divergent requirements for certification among these standards. Based on significant commonalities, we have developed a common process model for the development, assessment and certification of safety critical programmable electronic systems which could be acceptable in the framework of each standard in consideration. The proposed model contains a system development and a safety assessment process which rationalises and unifies the common requirements among the standards in these areas. In addition, it defines a common evolutionary process for the development of the system's safety case. The safety case process determines how the evidence produced in the progression of safety assessment can be structured in order to form an overall convincing argument about the safety of the system. We conclude that it is possible to use this model as the basis of a generic method for the certification of systems across the transportation sector and outline a suitable approach to such certification.

A harmonised model for safety assessment and certification of safety-critical systems in the transportation industries

This paper describes a model for the assessment and certification of safety-critical programmable electronic systems in the transportation industries. The proposed model is founded on the significant commonalities between emerging international safety-related standards in the automotive, railway and aerospace industries. It contains a system development and a safety assessment process which rationalise and unify the common requirements among the standards in these areas. In addition, it defines an evolutionary process for the development of the system's safety case. The safety case process shows how the evidence produced in the progression of safety assessment can be structured in order to form an overall argument about the safety of the system. We conclude that it is possible to use this model as the basis of a generic approach to the certification of systems across the transportation sector.